MP Econ Issue 6: Why Is Housing So Expensive in Urban China?

America’s Silicon Valley and China’s Silicon Delta share a serious problem: housing affordability. For San Francisco, the astronomical price of housing may increasingly deter the very human capital on which the tech hub so desperately relies.  

Halfway around the world, Shenzhen too has transformed from a sleepy swamp that was once populated by farmers into a gleaming but exorbitantly expensive tech hub. Between 2014 to 2020, Shenzhen property prices have more than doubled, far exceeding the ~60% increase in San Francisco during the same period. In fact, Shenzhen’s average apartment price has risen so much that some are finding it cheaper to live in Hong Kong, one of the most expensive property markets in the world.  

But the similarities between the two cities end there. San Francisco’s housing problem seems mainly attributable to regulatory restrictions on building more housing supply. The issue for Shenzhen is more fundamental: the city government limits the allocation of residential land because it owns all of it.  

Herein lies the rub: at the heart of China’s housing booms and busts is a distorted land allocation system that underpins the entire property market. Shenzhen, then, reflects a microcosm of a system that often drives housing prices into the stratosphere, beyond the reach of affordability. 

Abundance of Commercial Land 

Demand factors such as population inflows certainly contributed to Shenzhen’s housing shortage. But an important factor is the local government’s land supply policy that tends to favor commercial development over households. Between 2008-2018, despite seeing more than five million people move into the city, the Shenzhen government supplied almost three times more land for commercial use than for residential use.  

Shenzhen’s pro-business bias in land allocation can be illustrated by the case of Huawei, the Chinese tech giant that has garnered quite a global reputation. Headquartered in Shenzhen, Huawei in 2012 decided to build an Apple Park-style new campus in Dongguan, a manufacturing hub one-hour away from Shenzhen. Huawei’s reason for relocating? High land prices.  

This is tantamount to Amazon leaving Seattle because it couldn’t afford to stay in the city. And as expected, the Huawei announcement shocked Shenzhen officials and spurred them into action. The Shenzhen government eventually reached a deal with Huawei to provide “support and services” to the company’s satisfaction for it to keep its headquarters in the city.  

That scarring experience may have led Shenzhen officials to overlearn the importance of having affordable commercial land to attract businesses. By 2017, the city’s commercial land supply was more than 40 times the residential land supply (see Figure 1).  

Figure 1. Shenzhen’s Commercial Land Supply Far Outstrips Residential Land 

Source: Wind and MacroPolo. 

This strongly biased land policy has had negative effects on Shenzhen’s income distribution and overall growth. Since 2014, the year Shenzhen sharply adjusted its policy to favor commercial land, the city has seen below average GDP growth. Meanwhile, the city’s fiscal revenue and corporate profit growth both exceeded the national average by a large margin.  

But that has come at the expense of household income growth, which has been below the national average (see Figure 2). This also doesn’t help China’s low consumption problem, since consumers have less disposable income. And to the extent they have money to spend, it is probably concentrated on saving for one major purchase: a home. 

Figure 2. Shenzhen Government and Businesses Grow at the Expense of Household Income 

Note: Positive value indicates Shenzhen growing faster than national average. Corporate profit is proxied by corporate income tax. Shenzhen’s corporate profits dipped in 2018 due to the US-China trade war. 

Source: Wind and MacroPolo. 

Scarcity of Residential Land

Shenzhen’s problem is not unique and is symptomatic of the national property market. China’s urban housing boom is well known. Since 2015, residential property prices have appreciated by more than 50% in China’s largest cities. Over the past decade, average residential land supply per new resident in the top 10 cities is only 230 square feet—little more than the size of a typical hotel room—or less than 60% of the average per capita residential space in China.  

The commercial property market, on the other hand, is just the opposite. That market seems to be in a state of perpetual oversupply, with average office vacancy rates hovering around 20% over the last decade. 

It is puzzling that local governments tend to flood the market with commercial land when there is clearly a premium for selling residential land (see Figure 3). While it is tempting to attribute the proximate cause to the Chinese state’s monopoly over land ownership, it actually has more to do with local governments’ incentives for allocating land.  

Figure 3. Stronger Demand for Residential Land  

Source: Wind.  

That’s because the main prerogative for local governments is to accrue benefits, usually in the form of fiscal revenue, to themselves. Therefore, if they can attract more businesses into their jurisdictions, and if those businesses become profitable, then the local government will collect more corporate taxes.  

With ever increasing competition between cities, local governments face intensifying pressure to vie for businesses. So they have resorted to one of the largest assets at their disposal, commercial land, as a major incentive. The people and the job opportunities, it is assumed, will follow. 

At the same time, residential land supply is deliberately kept scarce so the government can actually make money on residential land sales. In effect, residential land sales serve as a cross-subsidy on local governments’ pro-business land policy that sells commercial land cheaply. Ultimately, local governments get revenue from selling residential land at a premium price and get revenue from businesses that locate there—a twofer that primarily serves the local governments’ interests, not households.  

Property Tax to the Rescue?

Beijing has taken notice. The central government has called on major cities to increase residential land. Since 2019, Shenzhen has supplied more residential land than in the previous decade combined.  

Yet it is too early to declare victory. The central government can mandate local governments to increase residential land supply, but a long-term solution requires more fundamental changes to land ownership. That isn’t likely anytime soon because land is a financial asset for local governments, which means it directly affects local financial stability. If land is privatized, or its allocation and conversion become more market-based, that amounts to local governments losing their largest source of revenue.  

Given that local governments currently have $10 trillion in debt, and land sales being the most crucial and reliable source of income for debt repayment, any drastic changes to land ownership will seriously raise the risk of local government defaults. That caution is reflected in the absence of notable changes in the 2019 land law revisions.  

Meanwhile, Beijing is pushing hard for implementing the long-delayed property tax, which in theory is meant to reshape the land market. Property tax could provide an alternative source of revenue for municipal governments, which should reduce their reliance on land sales. By changing that incentive, local governments may be more inclined to naturally supply more residential land.  

So goes the thinking anyway. But the property tax in reality may have the opposite effect than intended. That is, a progressive property tax would make it more expensive for wealthy individuals to own multiple units and therefore reduce demand without affecting much change on the supply side.  

That’s because the property tax is unlikely to come anywhere close to offsetting the revenue loss that would result from selling less land. With a 45% marginal tax rate for the top income bracket, personal income tax revenue is equal to around 1% of China’s GDP. Given that Beijing has openly stated that average households will be exempt from the property tax, it seems unlikely to generate more revenue than the income tax.  

In contrast, annual land sales are currently on the order of more than 7% of GDP. Even after deducting for cost, it is safe to say that land sales generate revenue equal to at least 3% of China’s GDP. In fact, experiences in Shanghai and Chongqing—where each city has imposed a property tax for a decade—are illustrative. Revenue from property tax is only around 10% of land sales in Shanghai and less than 5% for Chongqing. Unless the property tax can fill that gaping hole for local governments, there is little hope that a property tax will significantly alter China’s land allocations.  

It is no surprise, then, why local governments have been lukewarm on the property tax. This lackluster interest will make it much more challenging for the central government to nationalize the tax. Even though the central government can mandate the tax, it is ultimately up to local governments to tailor and implement the tax based on local conditions. In the near-term, not much uptake should be expected on the tax.  

Fundamentally, China’s land supply distortions are rooted in local politics. It’s a system that has served local governments well at the expense of households, which means it has strong status quo inertia and is resistant to reform.  

So the hope of sweeping changes on land policy should be tempered. And that means expensive urban housing in China is here to stay, at least for the foreseeable future.